“…They are thus challenging for some of the most influential models of spoken word recognition (Gaskell & Marslen-Wilson, 1997; Marslen-Wilson & Warren, 1994; Marslen-Wilson & Welsh, 1978; Marslen-Wilson, 1990; McClelland & Elman, 1986; Norris, 1994) that code for the precise order of segments, and assume that the phonological form of words consists of an ordered sequence of sounds. As we discussed in our preceding papers (Dufour & Grainger, 2019, 2020), the TISK model (Hannagan et al, 2013; see You & Magnuson, 2018, for a more recent implementation) is at present the sole model of spoken word recognition that can account for transposed-phoneme effects 1 . TISK is an interactive-activation model similar to the TRACE model (McClelland & Elman, 1986), but it replaces the position-dependent units in TRACE by both a set of position-independent phoneme units (see Bowers et al, 2016, for evidence for a role for position-invariant phonemes in spoken word recognition) and a set of open-diphone units that represent ordered sequences of contiguous and noncontiguous phonemes (cf.…”